Patent US5498943A successfully refreshed from Google Patents.
US5498943A
Expired / Lapsed
US
⏱️ 0.0 years remaining
Feedback control device
Assignee: Fujitsu Ltd
•
Filing: 1993-08-30
•
Granted: 1996-03-12
•
Expires: 2013-08-30
Abstract
Abstract A feedback control device at a certain control time, predicts a feedback quantity at the next control time on the basis of a feedback quantity fed back from a controlled object and then performs a control operation based on the predicted feedback quantity. The feedback control device includes a predictive control unit for producing such an actuating signal as to decrease a deviation between the predicted feedback quantity and a desired value. The actuating signal is used to control the controlled object. This eliminates the detrimental effects of time delays associated with the controlled object or the control device, thus ensuring good control. The use of the predictive control unit permits the controlled quantity from the controlled object to converge to the desired value monotonically and quickly. The predictive control unit may use a layered neural network having an input layer supplied with a feedback quantity and an input value corresponding to an actuating signal, and an output layer outputting a predicted feedback quantity. The deviation is back propagated from the output layer to the input layer by a relaxation algorithm in order to update the input value. The updated input value is applied to an actuating unit as an actuating signal, thereby controlling the controlled object.
Inventors
- 👤 Takashi Kimoto
- 👤 Daiki Masumoto
- 👤 Hiroshi Yamakawa
- 👤 Shigemi Nagata
Bibliographic Details
Application Number
—
Jurisdiction
US
Data Source
google_patents
Last Synchronized
2026-10-30 09:59
CPC Classifications
G; G05; G05B; G05B13/00; G05B13/02; G05B13/0265; G05B13/027
Claims Analysis
Total Claims: 15
Claim 1
Independent Claim
109 words
A feedback control device comprising: converting means for converting a controlled quantity output from a controlled object to a feedback quantity at a first control time comparable with a desired value; predictive control means comprising a layered neural network and responsive to said feedback quantity, for generating a predicted feedback quantity at a second control time, back propagating a difference between said predicted feedback quantity and said desired value toward an input side of said layered neural network using a relaxation algorithm, and producing an actuating signal that decreases the difference; and actuating means responsive to said actuating signal for providing a corresponding control input to said controlled object.
Claim 2
Dependent Claim
109 words
A feedback control device comprising: converting means for converting a controlled quantity output from a controlled object to a feedback quantity at a first control time comparable with a desired value; predictive control means comprising a layered neural network and responsive to said feedback quantity, for generating a predicted feedback quantity at a second control time, back propagating a difference between said predicted feedback quantity and said desired value toward an input side of said layered neural network using a relaxation algorithm, and producing an actuating signal that decreases the difference; and actuating means responsive to said actuating signal for providing a corresponding control input to said controlled object.
Claim 3
Dependent Claim
30 words
A feedback control device according to claim 1, wherein said controlled quantity, said feedback quantity, said desired value, said actuating signal, and said control input are each a vector quantity.
Claim 4
Dependent Claim
31 words
A feedback control device according to claim 1, wherein said desired value is constant regardless of said first and second control times, and said control device performs a constant value control.
Claim 5
Dependent Claim
22 words
A feedback control device according to claim 1, wherein said desired value is programmable, and said control device performs a programed control.
Claim 6
Dependent Claim
22 words
A feedback control device according to claim 1, wherein said desired value changes arbitrarily, and said control device performs a follow-up control.
Claim 7
Dependent Claim
38 words
A feedback control device according to claim 1, wherein said controlled object outputs a plurality of controlled quantities, and said converting means provides said predictive control means with a plurality of feedback quantities corresponding to said controlled quantities.
Claim 8
Dependent Claim
38 words
A feedback control device according to claim 1, wherein said predictive control means outputs a plurality of actuating signals, and said actuating means provides said controlled object with a plurality of control inputs corresponding to said actuating signals.
Claim 9
Dependent Claim
85 words
A feedback control device according to claim 1, wherein said layered neural network comprises a first input-layer unit supplied with said feedback quantity, a second input-layer unit supplied with an input value corresponding to said actuating signal, and an output-layer unit for outputting said predicted feedback quantity; updates said input value corresponding to said actuating signal to reduce the difference between said predicted feedback quantity predicted by said layered neural network and said desired value; and applies the updated input value to said actuating means.
Claim 10
Dependent Claim
24 words
A feedback control device according to claim 8, wherein a weight of coupling between units of said layered neural network is changed by learning.
Claim 11
Dependent Claim
163 words
A feedback control device according to claim 8, wherein said predictive control means further comprises: input value setting means for applying said input value corresponding to said actuating signal to said second input layer unit; initial value setting means for applying an initial value of said input value corresponding to said actuating signal to said input value setting means; error back propagation means for back propagating the difference as an error from said output-layer unit, to said second input-layer unit using said relaxation algorithm to thereby obtain an amount of correction for said input value to said second input-layer unit; input value update means responsive to said amount of correction for updating said input value corresponding to said actuating signal and applying the updated input value to said input value setting means; and input value fetch means for fetching the updated input value from said input value setting means and applying the update input value to said actuating means as an actuating signal.
Claim 12
Dependent Claim
55 words
A feedback control device according to claim 10, wherein said predictive control means further comprises end determination means for determining whether an end condition is met and outputting an update end signal to said error back propagation means, said input value update means and said input value fetch means when said end condition is met.
Claim 13
Dependent Claim
40 words
A feedback control device according to claim 10, wherein said error back propagation means obtains an amount of correction which permits said input value corresponding to said actuating signal to be within an allowable input range of said actuating means.
Claim 14
Dependent Claim
85 words
A predictive controller with a layered neural network, comprising: first input-layer means for receiving a first quantity of a first control time; second input-layer means for receiving a second quantity; output-layer means for outputting a predicted quantity corresponding to said first quantity at a second control time; and updating means for back propagating a difference between said predicted quantity and a desired value of said first quantity from said output-layer means to said second input-layer means, thereby updating said second quantity to reduce the difference.
Claim 15
Dependent Claim
85 words
A predictive controller with a layered neural network, comprising: first input-layer means for receiving a first quantity of a first control time; second input-layer means for receiving a second quantity; output-layer means for outputting a predicted quantity corresponding to said first quantity at a second control time; and updating means for back propagating a difference between said predicted quantity and a desired value of said first quantity from said output-layer means to said second input-layer means, thereby updating said second quantity to reduce the difference.
References Cited (Backward Citations)
USPTO Annuity & Maintenance Windows
| Annuity Stage | Window Opens | Payment Deadline | Surcharge Period Until | Estimated Fee | Status |
|---|---|---|---|---|---|
| 3.5 Years (Year 4) | 1999-03-12 | 1999-09-11 | 2000-03-12 | $2000.00 | Window Closed / Lapsed |
| 7.5 Years (Year 8) | 2003-03-12 | 2003-09-11 | 2004-03-12 | $3760.00 | Window Closed / Lapsed |
| 11.5 Years (Year 12) | 2007-03-12 | 2007-09-11 | 2008-03-12 | $7700.00 | Window Closed / Lapsed |